Reproduce “Physics Informed Random Feature Neural Networks for Solving PDEs” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Chi-An Chen, Chunyang Liao, Ming Zhong

The paper presents PIRFN, a physics-informed PDE solver that replaces trainable hidden layers with fixed, sampled features and learns only output coefficients. It separates spatial and temporal features, combining them to represent different scales while enforcing equations and conditions. For periodic transport with c=12, PIRFN reached 5.30×10^-3 held-out relative L^2 error versus 1.00 for tested PINN, on that benchmark.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.